Dashboard Charts - Event Centre



This chart is only visible to Assurance users.


To access the Event Centre, select a dashboard row and then click the Summary button at the top of the page:




When navigating to the Event Centre, the Assurance Charts window will show whether there is a stopped or ongoing event. Clicking Event Centre will take the user to the charts.




Overview




Automated Alert Triage Bar



This bar shows details of the event including:


  • Event Status - whether the event is ongoing or has stopped. 
  • Event duration - the start and, if applicable, end date and time of the event, and its duration.
  • Latest Analytic Information - which percentage of active analytics are alerting and the number of inactive analytcs.
  • Why did this event stop? - a reason for the event ending.



Along the left, 3 tabs are available which display data for the event.


  • Causes - lists the possible causes for the event and the analytics which identified the cause. This tab is discussed in greater detail below.
  • Suppliers - lists all known suppliers involved in the event and the analytics which identified the supplier.
  • Analytics - lists all the analytics in the dashboard row.



Causes


Potential causes of the event will appear in the Causes tab in alphabetical order. It will show how many analytics identified this as the potential cause, as well as what percentage this is of the total alerting analytics.




There are six potential causes which can be identified:


  • Path Change - there has been a significant change in the network path taken from the Digital User to the target application. This change of path has caused the quality score to drop and has caused an event to be raised.
  • Application Congestion - the application has been identified as an impairer rather than the network delivering the service from the application.
  • Application Error - tither there is a misconfiguration in the application or Digital User or the application is unable to serve content; this is causing error responses to be observed.
  • Network Congestion - congestion in the network has been identified as the cause of impairment. Such congestion could manifest as an increase in network delay only or an increase in network delay and observed packet loss.
  • Network Configuration - either there is a misconfiguration in the network or there is a traffic policing configuration that is causing an increase in packet loss but without any increase in delay behaviour.
  • Unknown- the Analytics Cloud has been unable to determine the underlying cause of the analytic event. This can be due to:
    • Insufficient Acute Diagnostic Measurements - the Digital User has the capability and the switch is on, however insufficient Acute Diagnostic measurements are available.  This is likely a Digital User issue, or a comms issue, for example between the Digital User and Analytics Cloud.
    • Insufficient History - the Analytics Cloud is still learning the “normal behaviour” of this analytic; there are not enough data points in the distribution used during the analysis.
    • Insufficient Scores - analytics stopped alerting before enough data was gathered to perform analysis.
    • Diagnosis Timeout - the amount of time the Analytics Cloud is prepared to wait for sufficient scores to be available in order to perform its processing on an open event has been exceeded.



Selecting a potential cause from this tab will display details of the alert in the main display area. If available, the two top impairments will appear below the potential cause name.




This will also show the contributing analytics alphabetically in the Causes tab. Selecting an analytic from the list will display more details in the main display area.

 



Analytic Details


This tab is available when an analytic is selected from the Causes, Suppliers or Analytics tabs.




Supply Chain - shows the path through the network from the DU to the target. If available, primary and secondary impairers will be highlighted. There is a key below the chart which explains what the symbols and colours mean.


Possible Primary Diagnosis - shows the possible cause. If available, primary and secondary impairers information will show here.



Quality History


This tab enables the individual analytic quality scores to be investigated across the event timeline.




The Quality History Chart shows the Quality Score for the analytic plotted along the chart. The configured alert threshold in green and the event and alert thresholds in red.


Hovering over the chart will display information about the analytic behaviour during that time period. The selection can be locked, so it no longer follows cursor movement, by clicking on the desired point. It can be unlocked by clicking again.


The window shows the date and time selected, the quality score for the dashboard row aggregate and for the selected analytic. The event status is also displayed next to the analytic. At the bottom the number of alerting analytics and inactive analytics contributing to the event are displayed.




The two red bars below the chart depict the Automated Alerts Triage and the Event Chart for the analytic.


Clicking on the Automated Alerts Triage bar will show the possible cause:




The Event Chart depicts when the Event Threshold was breached and when the event ended as a red bar. Hovering over this bar will show a text description of this data.